In 2019, ~5.8 million Americans suffered from Alzheimer?s disease (AD) of which ~5.6 million were >65 years of age. One in 10 Americans age >65 has AD. Although common and rare variants in >20 genes have been implicated in LOAD etiology, with APOE playing a major role, many more remain to be discovered. Clinical risk factors for AD include stroke, hypertension, T2D, obesity, and dyslipidemia all of which are known to have a major genetic component contributing to their etiology. However, it is currently unknown if the same variants cause AD and another risk factor (pleiotropy), e. g. T2D, or if these effects are due to mediation. Pleiotropy, although an important phenomenon in genetic etiology, has not been adequately studied and methods are limited to detect pleiotropy for rare and imputed variants. Through the parent grant we have developed a framework to study pleiotropy which can be applied to AD and its risk factors to bring about a better understanding of AD etiology. We now have the unique ability to incorporate in this project the study of the genetics of AD by not only leveraging our analysis framework but also data from the UK Biobank (discovery) and AD-specific (replication) data sets. We will analyze these datasets for cross phenotype associations and test these variants for pleiotropy using mediation analysis for AD and a number of comorbid phenotypes, that include type 2 diabetes (T2D), stroke, blood pressure (BP), adiposity, and blood lipids. The discovery dataset is the UK Biobank study, a population-based prospective study that has extensive genotype and phenotype data on ~500,000 subjects from the United Kingdom. This dataset was selected as the main resource for genotype and phenotype data because it is one of the largest population-based studies available to the scientific community. Due to the lack of AD cases at this point in the UK Biobank, we will apply a proxy family history phenotype. This approach was previously applied in the UK Biobank for a number of traits, including AD, with ~314,000 subjects with information on parental history of AD, 42,034 of whom reported at least one parent with AD. For our replication sample, we will be using four data sets: The National Institute on Aging Late-Onset Alzheimer Disease (NIALOAD); The Washington Heights?Inwood Columbia Aging Project (WHICAP), The Estudio Familiar de Influencia Genetica en Alzheimer (EFIGA), and The Alzheimer?s Disease Neuroimaging Initiative (ADNI). These data sets will provide us with WGS data for 22,924 individuals of which 10,373 are AD cases. The results from this study may help to elucidate the causal relationship between genetic variants for AD and comorbid phenotypes.